# SSVEP-019: A Comparison Study of Canonical Correlation Analysis Based Methods for Detecting Steady-State Visual Evoked Potentials

> Generated from a local paper card. Do not edit this page directly; edit the source card and rerun `vp run docs:generate`.

> Internal wiki boundary: paper-card evidence and user-provided PDF downloads are available for private reading. Extracted full-paper text is not published.

> Source: `library/paper_cards/SSVEP-019.md`

## Paper Access

* Internal PDF: <a href={"/papers/SSVEP-019.pdf"} download style={{ display: "inline-flex", alignItems: "center", justifyContent: "center", minHeight: "2.25rem", padding: "0.45rem 0.8rem", borderRadius: "6px", backgroundColor: "#047857", color: "#ffffff", fontWeight: 700, lineHeight: 1, textDecoration: "none", boxShadow: "0 1px 2px rgba(15, 23, 42, 0.22)" }}>Download Paper</a>
* DOI / official page: [10.1371/journal.pone.0140703](https://doi.org/10.1371/journal.pone.0140703)
* Open-access page: [Open access page](https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0140703)
* Deployment boundary: these PDF links are intended only for a private/protected internal wiki.

## SSVEP-019: A Comparison Study of Canonical Correlation Analysis Based Methods for Detecting Steady-State Visual Evoked Potentials

## Metadata

* ID: SSVEP-019
* Title: A Comparison Study of Canonical Correlation Analysis Based Methods for Detecting Steady-State Visual Evoked Potentials
* Year: 2015
* DOI / URL: 10.1371/journal.pone.0140703
* Local PDF: see Paper Access section above
* Text artifact: local-only path withheld from docs site
* Review status: `extracted`

## Study Type

* Track: SSVEP
* Task: compare CCA-based SSVEP detection methods
* Participants or dataset: 10 healthy subjects, 12-class SSVEP dataset
* Device/electrode setup: EEG recorded at 2048 Hz; detailed montage needs confirmation
* Protocol/task: simulated online BCI experiment with 12 targets and 15 blocks

## Methods

* Signal processing or analysis: standard CCA and extended CCA methods using calibration data
* Metrics: classification accuracy and simulated ITR
* Online/offline: simulated online/offline analysis

## Key Results

* Extended CCA methods using individual calibration data improved detection performance relative to standard CCA.
* The paper provides the comparison layer between original CCA and later template/filter-bank methods.

## Limitations

* Uses a fixed 12-class visual target layout.
* The results do not test moving objects, AR overlays, or real robot manipulation.
* Calibration-data requirements must be separated from calibration-free claims.

## Relevance To Current Review

* Important for preventing an oversimplified "CCA -> FBCCA -> TRCA" narrative.
* Helps explain why calibration and individual-template methods matter.

## Evidence Status

| Claim | Status | Evidence Note |
| --- | --- | --- |
| CCA-based SSVEP methods include both calibration-free and calibration-dependent variants. | verified | Abstract contrasts standard CCA with extended methods incorporating individual calibration data. |
| Individual calibration data can improve CCA-based SSVEP detection. | verified | Abstract states calibration data significantly improved detection performance. |
| Calibration-dependent CCA should be used in SAH-BRI-Grasp without setup cost analysis. | needs confirmation | Calibration burden is a system-design tradeoff. |

## Open Questions

* Which exact extended CCA variant should be compared against FBCCA/TRCA in Exp1?
* Should calibration burden be reported as a primary metric in dynamic-object SSVEP tests?